UTS IF540 MACHINE LEARNING
¶

Covid Detection Based On Symtomps
¶

Linear Discriminant Analysis
¶

Semester Ganjil 2022/2023
¶

Dataset yang dipakai:¶

https://www.kaggle.com/datasets/hemanthhari/symptoms-and-covid-presence?resource=download

Referensi yang dipakai:¶

https://www.kaggle.com/code/midouazerty/symptoms-covid-19-using-7-machine-learning-98 https://www.kaggle.com/code/meesalasaidhanush/symptoms-and-covid-presence-99-acc https://www.kaggle.com/code/dzuljalali/covid-19-classification-using-svm-svc

Hasil kerja¶

Library¶

In [126]:
# import the Libraries
import numpy as np
import pandas as pd

# visualisasi data
import matplotlib
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline

# dataprep
from dataprep.eda import *
from dataprep.eda.missing import plot_missing
from dataprep.eda import plot_correlation

# splitting the dataset into train set and test set
from sklearn.model_selection import train_test_split
from sklearn import metrics
from sklearn.metrics import accuracy_score,confusion_matrix

# LDA
## feature scaling
from sklearn.preprocessing import StandardScaler

## import LDA model
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis as LDA

## import the Logistic Regression model from sklearn using the 2 variances with the help of LDA
from sklearn.linear_model import LogisticRegression

Import Dataset¶

In [127]:
# import the dataset
df = pd.read_csv('Covid Dataset.csv')

Data Preprocessing¶

Data Information¶
In [128]:
df.head(5)
Out[128]:
Breathing Problem Fever Dry Cough Sore throat Running Nose Asthma Chronic Lung Disease Headache Heart Disease Diabetes ... Fatigue Gastrointestinal Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places Wearing Masks Sanitization from Market COVID-19
0 Yes Yes Yes Yes Yes No No No No Yes ... Yes Yes No Yes No Yes Yes No No Yes
1 Yes Yes Yes Yes No Yes Yes Yes No No ... Yes No No No Yes Yes No No No Yes
2 Yes Yes Yes Yes Yes Yes Yes Yes No Yes ... Yes Yes Yes No No No No No No Yes
3 Yes Yes Yes No No Yes No No Yes Yes ... No No Yes No Yes Yes No No No Yes
4 Yes Yes Yes Yes Yes No Yes Yes Yes Yes ... No Yes No Yes No Yes No No No Yes

5 rows × 21 columns

In [129]:
df.describe()
Out[129]:
Breathing Problem Fever Dry Cough Sore throat Running Nose Asthma Chronic Lung Disease Headache Heart Disease Diabetes ... Fatigue Gastrointestinal Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places Wearing Masks Sanitization from Market COVID-19
count 5434 5434 5434 5434 5434 5434 5434 5434 5434 5434 ... 5434 5434 5434 5434 5434 5434 5434 5434 5434 5434
unique 2 2 2 2 2 2 2 2 2 2 ... 2 2 2 2 2 2 2 1 1 2
top Yes Yes Yes Yes Yes No No Yes No No ... Yes No No Yes No Yes No No No Yes
freq 3620 4273 4307 3953 2952 2920 2869 2736 2911 2846 ... 2821 2883 2983 2726 2924 2820 3172 5434 5434 4383

4 rows × 21 columns

In [130]:
df.columns
Out[130]:
Index(['Breathing Problem', 'Fever', 'Dry Cough', 'Sore throat',
       'Running Nose', 'Asthma', 'Chronic Lung Disease', 'Headache',
       'Heart Disease', 'Diabetes', 'Hyper Tension', 'Fatigue ',
       'Gastrointestinal ', 'Abroad travel', 'Contact with COVID Patient',
       'Attended Large Gathering', 'Visited Public Exposed Places',
       'Family working in Public Exposed Places', 'Wearing Masks',
       'Sanitization from Market', 'COVID-19'],
      dtype='object')
In [131]:
plot_missing(df)

  0%|          | 0/230 [00:00<?, ?it/s]
C:\Users\hp\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\dask\core.py:119: RuntimeWarning: invalid value encountered in divide
  return func(*(_execute_task(a, cache) for a in args))
Out[131]:
DataPrep.EDA Report

Missing Statistics

Missing Cells0
Missing Cells (%)0.0%
Missing Columns0
Missing Rows0
Avg Missing Cells per Column0.0
Avg Missing Cells per Row0.0
'height': 500
Height of the plot
'width': 500
Width of the plot
'spectrum.bins': 20
Number of bins
'height': 500
Height of the plot
'width': 500
Width of the plot
'height': 500
Height of the plot
'width': 500
Width of the plot
'height': 500
Height of the plot
'width': 500
Width of the plot
In [132]:
# create a table with data missing 
missing_values=df.isnull().sum() # missing values

percent_missing = df.isnull().sum()/df.shape[0]*100 # missing value %

value = {
    'missing_values ':missing_values,
    'percent_missing %':percent_missing  
}
frame=pd.DataFrame(value)
frame
Out[132]:
missing_values percent_missing %
Breathing Problem 0 0.0
Fever 0 0.0
Dry Cough 0 0.0
Sore throat 0 0.0
Running Nose 0 0.0
Asthma 0 0.0
Chronic Lung Disease 0 0.0
Headache 0 0.0
Heart Disease 0 0.0
Diabetes 0 0.0
Hyper Tension 0 0.0
Fatigue 0 0.0
Gastrointestinal 0 0.0
Abroad travel 0 0.0
Contact with COVID Patient 0 0.0
Attended Large Gathering 0 0.0
Visited Public Exposed Places 0 0.0
Family working in Public Exposed Places 0 0.0
Wearing Masks 0 0.0
Sanitization from Market 0 0.0
COVID-19 0 0.0
In [133]:
sns.countplot(x='COVID-19',data=df)
Out[133]:
<AxesSubplot:xlabel='COVID-19', ylabel='count'>
In [134]:
df["COVID-19"].value_counts().plot.pie(explode=[0.1,0.1],autopct='%1.1f%%',shadow=True)
plt.title('number of cases');
Breathing Problem¶
In [135]:
sns.countplot(x='Breathing Problem',data=df)
Out[135]:
<AxesSubplot:xlabel='Breathing Problem', ylabel='count'>
In [136]:
sns.countplot(x='Breathing Problem',hue='COVID-19',data=df)
Out[136]:
<AxesSubplot:xlabel='Breathing Problem', ylabel='count'>
Fever¶
In [137]:
sns.countplot(x='Fever',hue='COVID-19',data=df);
Dry Cough¶
In [138]:
sns.countplot(x='Dry Cough',hue='COVID-19',data=df)
Out[138]:
<AxesSubplot:xlabel='Dry Cough', ylabel='count'>
Sore Throat¶
In [139]:
sns.countplot(x='Sore throat',hue='COVID-19',data=df)
Out[139]:
<AxesSubplot:xlabel='Sore throat', ylabel='count'>
Feature Transformation¶
In [140]:
from sklearn.preprocessing import LabelEncoder
e=LabelEncoder()
In [141]:
df['Breathing Problem']=e.fit_transform(df['Breathing Problem'])
df['Fever']=e.fit_transform(df['Fever'])
df['Dry Cough']=e.fit_transform(df['Dry Cough'])
df['Sore throat']=e.fit_transform(df['Sore throat'])
df['Running Nose']=e.fit_transform(df['Running Nose'])
df['Asthma']=e.fit_transform(df['Asthma'])
df['Chronic Lung Disease']=e.fit_transform(df['Chronic Lung Disease'])
df['Headache']=e.fit_transform(df['Headache'])
df['Heart Disease']=e.fit_transform(df['Heart Disease'])
df['Diabetes']=e.fit_transform(df['Diabetes'])
df['Hyper Tension']=e.fit_transform(df['Hyper Tension'])
df['Fatigue ']=e.fit_transform(df['Fatigue '])
df['Gastrointestinal ']=e.fit_transform(df['Gastrointestinal '])
df['Abroad travel']=e.fit_transform(df['Abroad travel'])
df['Contact with COVID Patient']=e.fit_transform(df['Contact with COVID Patient'])
df['Attended Large Gathering']=e.fit_transform(df['Attended Large Gathering'])
df['Visited Public Exposed Places']=e.fit_transform(df['Visited Public Exposed Places'])
df['Family working in Public Exposed Places']=e.fit_transform(df['Family working in Public Exposed Places'])
df['Wearing Masks']=e.fit_transform(df['Wearing Masks'])
df['Sanitization from Market']=e.fit_transform(df['Sanitization from Market'])
df['COVID-19']=e.fit_transform(df['COVID-19'])
In [142]:
# print dataset again
df.head(5)
Out[142]:
Breathing Problem Fever Dry Cough Sore throat Running Nose Asthma Chronic Lung Disease Headache Heart Disease Diabetes ... Fatigue Gastrointestinal Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places Wearing Masks Sanitization from Market COVID-19
0 1 1 1 1 1 0 0 0 0 1 ... 1 1 0 1 0 1 1 0 0 1
1 1 1 1 1 0 1 1 1 0 0 ... 1 0 0 0 1 1 0 0 0 1
2 1 1 1 1 1 1 1 1 0 1 ... 1 1 1 0 0 0 0 0 0 1
3 1 1 1 0 0 1 0 0 1 1 ... 0 0 1 0 1 1 0 0 0 1
4 1 1 1 1 1 0 1 1 1 1 ... 0 1 0 1 0 1 0 0 0 1

5 rows × 21 columns

In [143]:
df.dtypes.value_counts()
Out[143]:
int32    21
dtype: int64
Information After Transformation¶
In [144]:
df.describe(include='all')
Out[144]:
Breathing Problem Fever Dry Cough Sore throat Running Nose Asthma Chronic Lung Disease Headache Heart Disease Diabetes ... Fatigue Gastrointestinal Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places Wearing Masks Sanitization from Market COVID-19
count 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 ... 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.000000 5434.0 5434.0 5434.000000
mean 0.666176 0.786345 0.792602 0.727457 0.543246 0.462643 0.472028 0.503497 0.464299 0.476261 ... 0.519139 0.469452 0.451049 0.501656 0.461907 0.518955 0.416268 0.0 0.0 0.806588
std 0.471621 0.409924 0.405480 0.445309 0.498172 0.498648 0.499263 0.500034 0.498770 0.499482 ... 0.499680 0.499112 0.497644 0.500043 0.498593 0.499687 0.492984 0.0 0.0 0.395009
min 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ... 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.0 0.0 0.000000
25% 0.000000 1.000000 1.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ... 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.0 0.0 1.000000
50% 1.000000 1.000000 1.000000 1.000000 1.000000 0.000000 0.000000 1.000000 0.000000 0.000000 ... 1.000000 0.000000 0.000000 1.000000 0.000000 1.000000 0.000000 0.0 0.0 1.000000
75% 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 ... 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 0.0 0.0 1.000000
max 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 ... 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 0.0 0.0 1.000000

8 rows × 21 columns

In [145]:
df.hist(figsize=(20,15));
Correlation Between Features¶
In [146]:
corr=df.corr()
corr.style.background_gradient(cmap='coolwarm',axis=None)
Out[146]:
  Breathing Problem Fever Dry Cough Sore throat Running Nose Asthma Chronic Lung Disease Headache Heart Disease Diabetes Hyper Tension Fatigue Gastrointestinal Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places Wearing Masks Sanitization from Market COVID-19
Breathing Problem 1.000000 0.089903 0.159562 0.303768 0.055190 0.075318 -0.098291 -0.062172 -0.073366 0.055427 0.045256 0.000561 -0.075390 0.117795 0.214634 0.200304 0.066688 0.018295 nan nan 0.443764
Fever 0.089903 1.000000 0.127580 0.322235 0.081758 0.073953 -0.025160 -0.035416 -0.031462 0.050286 0.079001 -0.060458 -0.008067 0.128726 0.164704 0.070490 0.002252 0.012102 nan nan 0.352891
Dry Cough 0.159562 0.127580 1.000000 0.213907 -0.030763 0.086843 -0.043664 -0.035912 0.047566 -0.006593 0.081989 -0.039909 0.008251 0.331418 0.128330 0.117963 0.086176 0.163102 nan nan 0.464292
Sore throat 0.303768 0.322235 0.213907 1.000000 0.039450 0.081377 -0.050440 -0.015971 0.002177 0.001938 0.042811 -0.023290 0.025886 0.205986 0.189251 0.216438 0.079055 0.104378 nan nan 0.502848
Running Nose 0.055190 0.081758 -0.030763 0.039450 1.000000 -0.022763 -0.014376 0.068479 -0.056750 0.042961 -0.020445 0.007026 -0.014673 0.034526 0.003776 0.061099 0.032568 -0.061323 nan nan -0.005657
Asthma 0.075318 0.073953 0.086843 0.081377 -0.022763 1.000000 -0.033771 0.037064 0.076783 -0.012060 0.017707 0.006564 0.101909 0.068286 0.005046 -0.044592 0.020941 -0.115679 nan nan 0.089930
Chronic Lung Disease -0.098291 -0.025160 -0.043664 -0.050440 -0.014376 -0.033771 1.000000 -0.050480 -0.039860 0.046789 -0.010331 -0.047655 -0.050333 -0.088854 -0.062482 -0.020548 -0.093049 0.038343 nan nan -0.056837
Headache -0.062172 -0.035416 -0.035912 -0.015971 0.068479 0.037064 -0.050480 1.000000 0.048471 0.032390 -0.207489 0.052035 0.097778 0.043589 -0.082101 -0.162992 -0.005790 -0.012625 nan nan -0.027793
Heart Disease -0.073366 -0.031462 0.047566 0.002177 -0.056750 0.076783 -0.039860 0.048471 1.000000 -0.032956 0.049139 -0.058925 0.004121 -0.020761 -0.025593 -0.045437 0.086169 0.035000 nan nan 0.027072
Diabetes 0.055427 0.050286 -0.006593 0.001938 0.042961 -0.012060 0.046789 0.032390 -0.032956 1.000000 0.042543 -0.043903 0.040651 0.039013 -0.085696 -0.061650 -0.078212 0.097696 nan nan 0.040627
Hyper Tension 0.045256 0.079001 0.081989 0.042811 -0.020445 0.017707 -0.010331 -0.207489 0.049139 0.042543 1.000000 -0.027605 -0.067972 -0.016382 0.027307 0.002911 0.019174 0.048152 nan nan 0.102575
Fatigue 0.000561 -0.060458 -0.039909 -0.023290 0.007026 0.006564 -0.047655 0.052035 -0.058925 -0.043903 -0.027605 1.000000 0.009356 -0.068401 -0.027383 -0.031058 -0.009562 -0.025623 nan nan -0.044188
Gastrointestinal -0.075390 -0.008067 0.008251 0.025886 -0.014673 0.101909 -0.050333 0.097778 0.004121 0.040651 -0.067972 0.009356 1.000000 0.099577 0.025277 -0.017251 -0.061885 -0.027603 nan nan -0.003367
Abroad travel 0.117795 0.128726 0.331418 0.205986 0.034526 0.068286 -0.088854 0.043589 -0.020761 0.039013 -0.016382 -0.068401 0.099577 1.000000 0.080210 0.113399 0.069609 0.143094 nan nan 0.443875
Contact with COVID Patient 0.214634 0.164704 0.128330 0.189251 0.003776 0.005046 -0.062482 -0.082101 -0.025593 -0.085696 0.027307 -0.027383 0.025277 0.080210 1.000000 0.234649 0.079800 0.006909 nan nan 0.357122
Attended Large Gathering 0.200304 0.070490 0.117963 0.216438 0.061099 -0.044592 -0.020548 -0.162992 -0.045437 -0.061650 0.002911 -0.031058 -0.017251 0.113399 0.234649 1.000000 0.083795 0.063776 nan nan 0.390145
Visited Public Exposed Places 0.066688 0.002252 0.086176 0.079055 0.032568 0.020941 -0.093049 -0.005790 0.086169 -0.078212 0.019174 -0.009562 -0.061885 0.069609 0.079800 0.083795 1.000000 0.028486 nan nan 0.119755
Family working in Public Exposed Places 0.018295 0.012102 0.163102 0.104378 -0.061323 -0.115679 0.038343 -0.012625 0.035000 0.097696 0.048152 -0.025623 -0.027603 0.143094 0.006909 0.063776 0.028486 1.000000 nan nan 0.160208
Wearing Masks nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan
Sanitization from Market nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan
COVID-19 0.443764 0.352891 0.464292 0.502848 -0.005657 0.089930 -0.056837 -0.027793 0.027072 0.040627 0.102575 -0.044188 -0.003367 0.443875 0.357122 0.390145 0.119755 0.160208 nan nan 1.000000
In [147]:
# beberapa data sama sekali tidak berkolerasi sama sekali contohnya
# Running Nose / Asthma /Chronic Lung Disease / Headache / Heart Disease / Diabetes / Fatigue / Gastrointestinal / Wearing Masks / Sanitization from Market
# maka kita akan melakukan drop kepada data - data tersebut

df=df.drop('Running Nose',axis=1)
df=df.drop('Chronic Lung Disease',axis=1)
df=df.drop('Headache',axis=1)
df=df.drop('Heart Disease',axis=1)
df=df.drop('Diabetes',axis=1)
df=df.drop('Gastrointestinal ',axis=1)
df=df.drop('Wearing Masks',axis=1)
df=df.drop('Sanitization from Market',axis=1)
df=df.drop('Asthma',axis=1)
df=df.drop('Fatigue ',axis=1)
In [148]:
corr=df.corr()
corr.style.background_gradient(cmap='coolwarm',axis=None)
Out[148]:
  Breathing Problem Fever Dry Cough Sore throat Hyper Tension Abroad travel Contact with COVID Patient Attended Large Gathering Visited Public Exposed Places Family working in Public Exposed Places COVID-19
Breathing Problem 1.000000 0.089903 0.159562 0.303768 0.045256 0.117795 0.214634 0.200304 0.066688 0.018295 0.443764
Fever 0.089903 1.000000 0.127580 0.322235 0.079001 0.128726 0.164704 0.070490 0.002252 0.012102 0.352891
Dry Cough 0.159562 0.127580 1.000000 0.213907 0.081989 0.331418 0.128330 0.117963 0.086176 0.163102 0.464292
Sore throat 0.303768 0.322235 0.213907 1.000000 0.042811 0.205986 0.189251 0.216438 0.079055 0.104378 0.502848
Hyper Tension 0.045256 0.079001 0.081989 0.042811 1.000000 -0.016382 0.027307 0.002911 0.019174 0.048152 0.102575
Abroad travel 0.117795 0.128726 0.331418 0.205986 -0.016382 1.000000 0.080210 0.113399 0.069609 0.143094 0.443875
Contact with COVID Patient 0.214634 0.164704 0.128330 0.189251 0.027307 0.080210 1.000000 0.234649 0.079800 0.006909 0.357122
Attended Large Gathering 0.200304 0.070490 0.117963 0.216438 0.002911 0.113399 0.234649 1.000000 0.083795 0.063776 0.390145
Visited Public Exposed Places 0.066688 0.002252 0.086176 0.079055 0.019174 0.069609 0.079800 0.083795 1.000000 0.028486 0.119755
Family working in Public Exposed Places 0.018295 0.012102 0.163102 0.104378 0.048152 0.143094 0.006909 0.063776 0.028486 1.000000 0.160208
COVID-19 0.443764 0.352891 0.464292 0.502848 0.102575 0.443875 0.357122 0.390145 0.119755 0.160208 1.000000
In [149]:
X=df.drop('COVID-19',axis=1)
y=df['COVID-19']

Machine Learning Model¶

LDA¶

In [150]:
# splitting the dataset into train set and test set
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, 
    random_state = 0)
    
# feature scaling
sc_X = StandardScaler()
X_train = sc_X.fit_transform(X_train)

X_test = sc_X.transform(X_test)

# initialize the LDA
lda = LDA(n_components = 1)

# fitting the LDA model
X_test = lda.fit_transform(X_test, y_test)
X_train = lda.transform(X_train)

LG=LogisticRegression(random_state=0)

# fit the Logistic Regression model
LG.fit(X_train, y_train)

# predict the Logistic regression model
y_pred = LG.predict(X_test)

Results¶

Confusion Matrix¶

In [151]:
from sklearn.metrics import confusion_matrix

# create the confusion matrix
cm = confusion_matrix(y_test, y_pred)

# print the confusion matrix
print(cm)

[[157  35]
 [ 18 877]]

Accuracy Score For Prediction¶

In [152]:
# import the accuracy score
from sklearn.metrics import accuracy_score

# print the accuracy score
print(accuracy_score(y_test, y_pred))

0.9512419503219871

Precision Score For Prediction¶

In [153]:
# import the classification report
from sklearn.metrics import classification_report

# print the classification report
print(classification_report(y_test, y_pred))

              precision    recall  f1-score   support

           0       0.90      0.82      0.86       192
           1       0.96      0.98      0.97       895

    accuracy                           0.95      1087
   macro avg       0.93      0.90      0.91      1087
weighted avg       0.95      0.95      0.95      1087

Save the notebook, then convert the notebook to html (by running the next code).¶

In [154]:
!jupyter nbconvert --to html "./LDA-Project" --output-dir="./"

Next step:¶

  • convert the generated html file to PDF using the online tool: https://www.sejda.com/html-to-pdf
  • choose the following settings:
    • Page size: One long page
    • Page Orientation: auto
    • Use print stylesheet
  • Submit your ipython notebook and PDF files

Markdown basics https://markdown-guide.readthedocs.io/en/latest/basics.html#